Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A computer-implemented method performed by at least one processor, the method comprising: training a neural network to process a first record of communications between a first service representative (SR) and a first individual during a first service session to generate one or more predicted survey scores that rate the first service session on one or more criteria, wherein the training is conducted using training data comprising, for each of one or more previous service sessions, a respective previous record of the previous service session and a respective set of one or more actual survey scores provided by a serviced individual to rate the previous service session on the one or more criteria; receiving, by at least one processor, a second record of communications between a second service representative (SR) and a second individual during a second service session; processing, using the trained neural network, the second record to determine one or more new predicted survey scores that rate the second service session, wherein the trained neural network comprises a language neural network model and an acoustic neural network model, and wherein the trained neural network comprises different neurons, each neuron corresponding to a respective criterion of the one or more criteria; associating, by at least one processor, the one or more new predicted survey scores with the second individual; and communicating, by at least one processor, the one or more new predicted survey scores for presentation through a user interface of a computing device.
This invention relates to a computer-implemented method for predicting customer satisfaction scores in service interactions using neural networks. The method addresses the challenge of evaluating service quality in real-time by analyzing communication data between service representatives and customers to predict survey scores without requiring manual post-interaction surveys. The method involves training a neural network using historical service session data, where each session includes communication records (e.g., text, audio) and corresponding actual survey scores provided by customers. The trained neural network processes new service interactions in real-time to generate predicted satisfaction scores for multiple criteria, such as responsiveness or problem resolution. The neural network combines language and acoustic models to analyze both textual and vocal communication patterns, with distinct neurons dedicated to each evaluation criterion. After processing a new interaction, the predicted scores are associated with the customer and displayed via a user interface, enabling immediate feedback for service representatives. This approach automates quality assessment, reducing reliance on post-interaction surveys and providing real-time insights to improve service delivery. The system enhances efficiency by leveraging machine learning to predict customer satisfaction dynamically.
2. The computer-implemented method of claim 1 , wherein: the second service session is an audio call between the SR and the individual; and the second record includes an audio record of at least a portion of the audio call.
This invention relates to a computer-implemented method for managing service sessions between a service representative (SR) and an individual, with a focus on recording and storing audio interactions. The method addresses the need to capture and retain audio data from service sessions, particularly audio calls, to improve service quality, compliance, or training. The method involves initiating a first service session between the SR and the individual, which may include interactions such as text, video, or other forms of communication. A second service session is then established as an audio call between the SR and the individual. During this audio call, at least a portion of the conversation is recorded, generating an audio record. This audio record is stored as part of a second record, which may also include metadata or other relevant data from the session. The method ensures that audio interactions are captured and preserved, allowing for later review, analysis, or compliance verification. The system may also support additional features, such as selective recording, encryption, or integration with other service management tools. The invention is particularly useful in customer service, healthcare, or other industries where audio communication is critical.
3. The computer-implemented method of claim 1 , wherein: the respective previous service session is an audio call; and the respective previous record includes an audio record of at least a portion of the audio call.
This invention relates to a computer-implemented method for managing service sessions, specifically focusing on audio calls. The method involves analyzing previous service sessions, such as audio calls, to improve future interactions. When a new service session is initiated, the system retrieves a previous service session record associated with the same user or context. For audio calls, this record includes an audio recording of at least part of the call. The system then uses this recorded audio to enhance the current session, such as by providing context, improving personalization, or assisting in decision-making. The method ensures that relevant historical data is leveraged to optimize user experience and service efficiency. The system may also process the audio record to extract key information, such as spoken content, speaker identification, or call duration, to support the current session. This approach helps maintain continuity in service interactions, particularly in scenarios where audio communication is critical, such as customer support, telemedicine, or virtual assistance. The invention aims to bridge gaps in communication by utilizing past interactions to inform and improve present ones.
4. The computer-implemented method of claim 1 , further comprising: updating the second individual requirements based on a metric; and selecting another service representative to interact with the second individual on a subsequent service session based in part on the updated requirements.
This invention relates to a computer-implemented method for optimizing service interactions between individuals and service representatives. The problem addressed is the inefficiency in matching individuals with service representatives, leading to suboptimal service experiences and resource allocation. The method involves analyzing interactions between an individual and a service representative during a service session to determine individual requirements. These requirements are used to select a service representative for a subsequent interaction. The method further includes updating these requirements based on a performance metric, such as customer satisfaction or resolution time, and using the updated requirements to select a different service representative for future interactions. This ensures continuous improvement in matching individuals with the most suitable representatives. The system may also involve tracking historical interaction data, applying machine learning models to predict optimal representative assignments, and dynamically adjusting selections based on real-time feedback. The goal is to enhance service quality, reduce resolution times, and improve overall customer satisfaction by leveraging data-driven decision-making in representative selection.
5. The computer-implemented method of claim 1 , further comprising: receiving a request for a service session from a third individual; identifying a fourth individual similar to the third individual; and connecting the third individual with another service representative, where the other service representative is selected, at least in part, based on metrics associated with the fourth individual.
This invention relates to a computer-implemented method for optimizing service sessions by dynamically matching individuals with service representatives based on similarity metrics. The method addresses the problem of inefficient service interactions by leveraging data-driven matching to improve outcomes. The method involves receiving a request for a service session from a third individual. The system then identifies a fourth individual who is similar to the third individual, using predefined similarity metrics such as demographic data, historical behavior, or interaction preferences. Based on this similarity, the system connects the third individual with a service representative who has demonstrated effective performance with the fourth individual or similar individuals. The selection of the service representative is influenced by metrics associated with the fourth individual, such as past satisfaction scores, resolution rates, or interaction efficiency. This approach ensures that service representatives with relevant experience or success in handling similar cases are assigned to the third individual, thereby enhancing the likelihood of a positive interaction. The method may also involve analyzing historical data to refine the similarity criteria and improve future matchings. The system dynamically adapts to user needs, optimizing resource allocation and service quality.
6. A non-transitory computer storage medium encoded with computer program instructions that when executed by one or more computers cause the one or more computers to perform operations comprising: training a neural network to process a first record of communications between a first service representative (SR) and a first individual during a first service session to generate one or more predicted survey scores that rate the first service session on one or more criteria, wherein the training is conducted using training data comprising, for each of one or more previous service sessions, a respective previous record of the previous service session and a respective set of one or more actual survey scores provided by a serviced individual to rate the previous service session on the one or more criteria; receiving, by at least one processor, a second record of communications between a second service representative (SR) and a second individual during a second service session; processing, using the trained neural network, the second record to determine one or more new predicted survey scores that rate the second service session, wherein the trained neural network comprises a language neural network model and an acoustic neural network model, and wherein the trained neural network comprises different neurons, each neuron corresponding to a respective criterion of the one or more criteria; associating, by at least one processor, the one or more new predicted survey scores with the second individual; and communicating, by at least one processor, the one or more new predicted survey scores for presentation through a user interface of a computing device.
This invention relates to a system for predicting customer satisfaction scores in service interactions using neural networks. The problem addressed is the need to evaluate service quality in real-time without relying solely on post-interaction surveys, which can be delayed or incomplete. The solution involves training a neural network to analyze communication records from service sessions and predict survey scores based on historical data. The system processes both language and acoustic data from interactions, using separate neural network models for each. The trained network generates predicted scores for specific criteria, such as responsiveness or problem resolution, by leveraging patterns from past interactions where actual survey scores were available. During a live service session, the system processes the ongoing communication in real-time, applies the trained neural network to predict satisfaction scores, and displays these predictions to relevant users. The system improves service quality monitoring by providing immediate feedback to service representatives and managers, enabling timely adjustments to enhance customer experience. The neural network's architecture includes distinct neurons for each evaluation criterion, allowing for granular performance assessment.
7. The non-transitory computer storage medium of claim 6 , wherein: the second service session is an audio call between the SR and the individual; and the second record includes an audio record of at least a portion of the audio call.
This invention relates to a system for managing service interactions, particularly focusing on recording and storing audio call data between a service representative (SR) and an individual. The system involves a non-transitory computer storage medium that stores instructions for handling service sessions, including a first service session where an individual interacts with a service system, and a second service session where the individual communicates with a service representative via an audio call. The system records at least a portion of this audio call and stores it as part of a second record associated with the interaction. This allows for documentation and review of the audio communication, which can be useful for quality assurance, training, or dispute resolution. The system may also link the second record to the first service session to provide a complete history of the interaction. The invention aims to improve service tracking and accountability by ensuring that audio interactions are properly documented and accessible for future reference.
8. The non-transitory computer storage medium of claim 6 , wherein: the respective previous service session is an audio call; and the respective previous record includes an audio record of at least a portion of the audio call.
This invention relates to a system for storing and retrieving records of previous service sessions, particularly audio calls, to enhance user experience in subsequent interactions. The system captures and stores audio records from past audio calls, allowing these records to be accessed and utilized during future service sessions. The stored audio records may include call transcripts, audio clips, or full call recordings, enabling context-aware responses or personalized interactions. The system is designed to improve efficiency and accuracy in service delivery by leveraging historical data from prior audio-based communications. The invention ensures that relevant information from previous calls is preserved and can be retrieved when needed, reducing redundancy and improving service quality. The stored records may be indexed or categorized to facilitate quick retrieval based on user identity, session type, or other metadata. This approach enhances continuity in service interactions, particularly in customer support, telehealth, or other domains where maintaining context across sessions is valuable. The system may integrate with existing communication platforms or operate as a standalone service, providing flexibility in deployment. The invention addresses the challenge of maintaining context in fragmented service interactions by preserving and utilizing audio records from prior sessions.
9. The non-transitory computer storage medium of claim 6 , wherein the operations further comprise: updating the second individual's requirements based on a metric; and selecting another service representative to interact with the second individual on a subsequent service session based in part on the updated requirements.
This invention relates to a system for dynamically improving customer service interactions by adapting service representative assignments based on evolving individual requirements. The problem addressed is the inefficiency of static assignment methods that fail to account for changing customer needs, leading to suboptimal service experiences. The system involves a computer-implemented method that tracks interactions between service representatives and individuals (e.g., customers) during service sessions. It analyzes these interactions to determine the individual's requirements, which may include preferences, communication styles, or specific needs. These requirements are stored in a data structure for future reference. A key feature is the ability to update an individual's requirements based on a metric derived from interaction data. This metric could reflect changes in the individual's behavior, feedback, or evolving needs over time. The system then uses these updated requirements to select a different service representative for subsequent interactions, ensuring better alignment between the individual's current needs and the representative's skills or attributes. The selection process considers multiple factors, including the updated requirements, the representative's profile, and historical interaction outcomes. This dynamic approach improves service quality by continuously refining assignments based on real-time data, rather than relying on fixed criteria. The system is particularly useful in customer service, healthcare, or any domain requiring personalized interactions.
10. The non-transitory computer storage medium of claim 6 , wherein the operations further comprise: receiving a request for a service session from a third individual; identifying a fourth individual similar to the third individual; and connecting the third individual with another service representative, where the other service representative is selected, at least in part, based on metrics associated with the fourth individual.
This invention relates to a system for optimizing service session routing in a customer support or interaction environment. The problem addressed is efficiently matching service representatives to individuals based on similarity metrics, ensuring better outcomes for both parties. The system uses data from prior interactions to identify patterns and preferences, improving the likelihood of successful service delivery. The system operates by storing interaction data from previous service sessions, including metrics such as satisfaction scores, resolution times, and interaction history. When a new request is received from an individual, the system analyzes this data to identify another individual with similar characteristics or needs. Based on the metrics associated with this similar individual, the system selects an appropriate service representative to handle the new request. This selection is designed to maximize compatibility and effectiveness, leveraging historical performance data to improve future interactions. The system may also track and update metrics in real-time, refining its matching algorithm over time. This ensures that the selection of service representatives remains accurate and aligned with evolving user needs. The overall goal is to enhance customer satisfaction and operational efficiency by dynamically optimizing the assignment of service representatives based on learned similarities between individuals.
11. A system comprising: one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising: training a neural network to process a first record of communications between a first service representative (SR) and a first individual during a first service session to generate one or more predicted survey scores that rate the first service session on one or more criteria, wherein the training is conducted using training data comprising, for each of one or more previous service sessions, a respective previous record of the previous service session and a respective set of one or more actual survey scores provided by a serviced individual to rate the previous service session on the one or more criteria; receiving, by at least one processor, a second record of communications between a second service representative (SR) and a second individual during a second service session; processing, using the trained neural network, the second record to determine one or more new predicted survey scores that rate the second service session, wherein the trained neural network comprises a language neural network model and an acoustic neural network model, and wherein the trained neural network comprises different neurons, each neuron corresponding to a respective criterion of the one or more criteria; associating, by at least one processor, the one or more new predicted survey scores with the second individual; and communicating, by at least one processor, the one or more new predicted survey scores for presentation through a user interface of a computing device.
This system operates in the domain of customer service quality assessment, addressing the challenge of evaluating service interactions in real-time to improve service quality and customer satisfaction. The system uses a neural network trained on historical service session data to predict survey scores for ongoing interactions. The training data includes records of past service sessions along with actual survey scores provided by customers, allowing the neural network to learn patterns that correlate communication content with customer satisfaction. The neural network processes both language and acoustic features of communications, using separate language and acoustic neural network models. Each criterion for evaluation (e.g., politeness, problem resolution) is represented by distinct neurons within the network, enabling multi-dimensional scoring. During a live service session, the system captures the communication between a service representative and a customer, processes it through the trained neural network, and generates predicted survey scores for each evaluation criterion. These scores are then associated with the customer and displayed through a user interface, providing real-time feedback to the service representative. This approach eliminates the need for post-session surveys, allowing immediate adjustments to improve service quality.
12. The system of claim 11 , wherein: the second service session is an audio call between the SR and the individual; and the second record includes an audio record of at least a portion of the audio call.
This invention relates to a system for managing service sessions between a service representative (SR) and an individual, particularly focusing on audio call interactions. The system captures and stores records of these sessions, including audio recordings of at least part of the audio calls. The system is designed to enhance service quality, compliance, and record-keeping by ensuring that interactions between the SR and the individual are documented. The audio recordings can be used for training, dispute resolution, or quality assurance purposes. The system may also integrate with other service sessions, such as text-based or video interactions, to provide a comprehensive record of the service encounter. The audio recording feature ensures that verbal agreements, instructions, or critical information exchanged during the call are preserved for future reference. The system may include additional features like metadata tagging, search functionality, or playback controls to facilitate efficient retrieval and analysis of the recorded sessions. This invention addresses the need for reliable documentation of audio-based service interactions in industries such as customer support, healthcare, or financial services, where accurate records of verbal communications are essential.
13. The system of claim 11 , wherein: the respective previous service session is an audio call; and the respective previous record includes an audio record of at least a portion of the audio call.
This invention relates to a system for managing and utilizing records of previous service sessions, particularly audio calls, to enhance user experience or system functionality. The system captures and stores audio records from past audio calls, allowing these records to be accessed, analyzed, or reused in subsequent interactions. The system may leverage these records to improve call routing, provide contextual assistance, or enable personalized services based on historical call data. By maintaining an audio record of at least a portion of each call, the system can reference past conversations to improve accuracy, efficiency, or user satisfaction in future sessions. The invention may be applied in customer service, telecommunication, or any domain where historical call data can enhance service delivery or automation. The system ensures that relevant portions of audio calls are preserved for future reference, enabling smarter decision-making or automated responses based on prior interactions.
14. The system of claim 11 , wherein the operations further comprise: updating the second individual's requirements based on a metric; and selecting another service representative to interact with the second individual on a subsequent service session based in part on the updated requirements.
This invention relates to a system for dynamically optimizing service representative assignments in customer service interactions. The problem addressed is the inefficiency in matching service representatives to individuals based on static or outdated requirements, leading to suboptimal service experiences. The system includes a processing unit that receives data from a first service session between a service representative and an individual. This data includes interaction details, performance metrics, and individual requirements. The system analyzes this data to determine the effectiveness of the interaction and updates the individual's requirements based on a metric, such as satisfaction scores, resolution time, or representative expertise. For subsequent service sessions, the system selects a different service representative by considering the updated requirements, ensuring better alignment between the individual's needs and the representative's skills. The selection may also factor in real-time availability, historical performance, and other contextual data. The system improves customer service by continuously refining representative assignments based on evolving individual needs and interaction outcomes, enhancing efficiency and satisfaction. The dynamic updating of requirements ensures that future interactions are tailored to the individual's current context, reducing repetition and improving resolution rates.
15. The system of claim 11 , wherein the operations further comprise: receiving a request for a service session from a third individual; identifying a fourth individual similar to the third individual; and connecting the third individual with another service representative, where the other service representative is selected, at least in part, based on metrics associated with the fourth individual.
This system operates in the domain of service session management, addressing the challenge of efficiently matching service representatives to individuals based on similarity metrics. The system receives a request for a service session from a third individual and identifies a fourth individual who is similar to the third individual. The similarity may be based on factors such as demographics, preferences, or historical interactions. The system then connects the third individual with another service representative, where the selection of the representative is influenced by metrics associated with the fourth individual. These metrics may include performance data, customer satisfaction scores, or expertise levels related to the fourth individual. The system ensures that the service representative chosen for the third individual is optimized based on the characteristics and historical data of the similar fourth individual, improving the likelihood of a successful and satisfactory interaction. The system may also include additional features such as real-time monitoring of service sessions, dynamic adjustment of representative assignments, and feedback mechanisms to refine future matching decisions. The overall goal is to enhance service efficiency and customer satisfaction by leveraging similarity-based representative selection.
Unknown
December 1, 2020
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.